At 12:25, on Mar 14, 2006, markallen@micron.com wrote:
I use DFM and MRTG to monitor several of filers. I've ran into issues with CPU stats for multi-processor systems. DFM and MRTG will report the "any value" versus the average value. I've had several discussions with Netapp tech support on this topic and there was never a clear direction on with metric to key off of. Tech support informed me that I should be looking at the CPU utilization with sysstat -m and key off of the AVG value.
...
My metrics using DFM and MRTG captures the "any" value. According to Netapp it is not the metric that should be captured. Is anyone else having these challenges and is their a fix in DFM to collect the correct metric on multi-proc systems?
Which MRTG are you using? I've found that querying SNMP for the CPU busy (.1.3.6.1.4.1.789.1.2.1.2) use will indeed graph the AVG CPU, which is the sane one.
I've never understood why most NetApp tools (all the way down to regular sysstat (except for -m)) report the ANY value.
Davin.
Davin Milun 14.03.2006 21:21:
At 12:25, on Mar 14, 2006, markallen@micron.com wrote:
I use DFM and MRTG to monitor several of filers. I've ran into issues with CPU stats for multi-processor systems. [...]
Which MRTG are you using? I've found that querying SNMP for the CPU busy (.1.3.6.1.4.1.789.1.2.1.2) use will indeed graph the AVG CPU, which is the sane one.
Actually, .1.3.6.1.4.1.789.1.2.1.2 is cpuBusyTime (at least the MIB for DOT 6.5.x says so) and calculating CPU utilization via delta(cpuBusyTime)*100/delta(cpuUpTime) on a FAS960 leaves us with a no-so-satisfying ANY value. See also Bug-IDs 141441 and 148982.
It would be better to have snmpd report a cpuTable with seperate data for all CPUs.
I've never understood why most NetApp tools (all the way down to regular sysstat (except for -m)) report the ANY value.
Marketing? "Look at that, 100% busy again! We need new filer heads" ...
-Cornelius
-
Cornelius Bartke -
Davin Milun